A Video Watermarking Based on 3-D Complex Wavelet

Abstract

A video watermarking embedding and detection algorithm is proposed in complex wavelet transform (CWT) domain. The host video sequence is firstly segmented into shots. Then each shot is projected onto 3D CWT domain. To achieve robustness while keeping imperceptibility, a perceptual mask derived from 3D complex wavelet coefficients is introduced to weight the watermarks and the results is added back to the complex wavelet coefficients. Finally the inverse 3D CWT is applied to obtain the watermarked video shots. The experimental results illustrate that the proposed algorithm has more advantages in remaining video qualities while keeping the same resistance to attacks over that in discrete wavelet transform (DWT) domain.